What is agentic commerce and when will AI agents buy fashion autonomously?
Agentic commerce spans a spectrum from assisted search to fully autonomous purchase. In 2026, agents recommend products widely but rarely complete checkout without human approval.

KEY TAKEAWAYS Summary by the editors
- Agentic commerce describes AI agents that handle shopping tasks from product discovery through to autonomous purchase, with varying levels of human oversight.
- Shopping searches on generative AI platforms grew 4,700 per cent between July 2024 and July 2025, according to McKinsey and Business of Fashion.
- Amazon disclosed that 300 million customers used its Rufus shopping assistant in 2025, with users 60 per cent more likely to complete a purchase.
- Only 7 per cent of fashion shoppers would allow an AI assistant to purchase with some degree of autonomy, according to a June 2026 survey of US and UK consumers.
What does agentic commerce mean?
Agentic commerce describes a form of e-commerce in which autonomous artificial intelligence agents independently execute purchasing and payment processes on behalf of users or organisations. A defining feature of agentic commerce is the delegation of end-to-end commercial activities to software agents. Based on these parameters, an agent can autonomously perform tasks including product discovery, price comparison, contract selection, order placement and payment execution.
AI shopping typically means a chatbot or recommendation engine that helps a human browse and decide. Agentic commerce goes further: an autonomous AI agent handles the entire transaction, from interpreting intent to comparing merchants to completing the purchase and managing delivery, without requiring the human to click through each step.
Where does agentic commerce sit on the AI shopping spectrum?
Agentic commerce is best understood as a spectrum rather than a binary state. At one end, AI assistants provide recommendations and comparisons but require human action at checkout. At the other, agents execute purchases with minimal or no oversight.
- Recommendation and discovery: AI assistants answer product questions, surface options and compare features, but humans make the final decision and complete checkout manually.
- Assisted purchase: Agents populate shopping carts, apply filters and prepare transactions, but require explicit human approval before completing payment.
- Constrained autonomy: Agents authorised to purchase within predefined parameters such as budget limits, specific retailers or repeat purchases of known items.
- Fully autonomous: Agents make independent purchasing decisions based on learned preferences, price monitoring and contextual triggers without transaction-by-transaction approval.
For low-risk, frequently repurchased staples, a shopper's preferred running sock or a regular white shirt reorder, the agent can be authorised to execute purchases autonomously within pre-defined parameters. Fashion purchases involving style, fit or occasion remain largely in the assisted categories, where human judgment remains central.
Which platforms deployed agentic shopping features in 2025 and 2026?
Amazon told its Q4 2025 earnings call that 300 million customers used its Rufus shopping assistant in 2025, and that Rufus users were about 60 per cent more likely to complete a purchase. In February 2024, Amazon launched its Rufus shopping assistant, which can help users find relevant products and answer any questions they might have about those products. In April 2025, Amazon also introduced its Buy for Me feature, which allows shoppers to buy products from third-party websites without leaving the Amazon Shopping app.
Stripe and OpenAI released the Agentic Commerce Protocol under the Apache 2.0 licence in September 2025, and ChatGPT Shopping launched with its first merchants. In March 2026, OpenAI scaled back Instant Checkout, its feature for buying products inside ChatGPT, after users researched products there but rarely completed a purchase.
Google rolled out new agentic shopping capabilities in November 2025. Shoppers can describe the product they are looking for in AI mode or Gemini, which can search through more than 50 billion product listings in Google's Shopping Graph. Google's AI will retrieve relevant product listings with prices, reviews and availability, and its AI agent can call nearby businesses to check if the item is in stock.
Perplexity Pro subscribers can click the Buy with Pro button to initiate an agent-driven checkout without leaving Perplexity. Perplexity's agent places the order using saved shipping and payment details, handling the transaction end to end. Checkout runs through PayPal and Venmo, and Perplexity takes no transaction fee.
Home improvement retailers have been some of the earliest adopters of agentic commerce, including Lowe's, which introduced its Mylow customer-facing AI-powered app in March 2025. Mylow offers agentic shopping so customers can check their order status and discover products but they also can ask how to paint their cabinets, what plants work in their postcode, what kitchen designs are trending in their area. Lowe's competitor The Home Depot introduced the Magic Apron AI assistant in March 2025, and the retailer has been fine-tuning the customer experience on an ongoing basis.
On 17 November 2026, Daydream, the AI-powered shopping platform built for fashion, unveiled the Daydream Agent, an autonomous personal stylist that can carry a styling request from occasion to outfit planning through purchase in a single conversation. Shoppers can also virtually try on the pieces and authorise the agent to complete a purchase with participating retail and brand partners.
What technical protocols enable agentic commerce?
Several technical protocols have been proposed to support interoperability among AI agents, merchants, payment providers and other participants in agentic-commerce systems. The Agentic Commerce Protocol is an open standard for programmatic commerce flows between buyers, AI agents and merchants. Google launched the Universal Commerce Protocol at NRF in January 2026, enabling AI agents to interact with merchant catalogues and complete purchases through a single open standard.
In 2025, Shopify continued investing on two fronts: empowering merchants to operate smarter through AI tools like Sidekick, its AI-enabled commerce assistant; and building the infrastructure for agentic commerce, including Universal Commerce Protocol, Catalog and Agentic Storefronts, to ensure Shopify is at the forefront of where commerce happens next. Shopify provides a suite of AI-enabled features integrated across the Shopify platform, including Sidekick, its AI-enabled commerce assistant, and Agentic Storefronts, which enables agent-assisted product discovery and checkout across a number of AI platforms.
For merchants, participation in agentic commerce may require products and services to be presented in structured, machine-readable formats to ensure discoverability and interoperability with autonomous agents.
How many consumers use AI agents for shopping today?
Shopping-related searches on generative AI platforms grew 4,700 per cent between July 2024 and July 2025, with AI supporting inspiration and product comparison, especially helpful in the fashion category where choice abounds. The McKinsey Agentic Commerce Consumer Survey, fielded in June 2026, surveyed approximately 21,000 global consumers across 11 markets. More than 50 per cent of global consumers now use AI for shopping.
A McKinsey survey found that 85 per cent of luxury consumers use a multipurpose AI assistant such as Google's AI Mode or Perplexity to support shopping decisions, with 52 per cent reporting frequent usage. Eighty-three per cent of luxury consumers report either high or very high satisfaction with AI shopping tools and only 4 per cent are dissatisfied.
The 2026 BoF-McKinsey State of Fashion found that 23 per cent of consumers primarily use generative AI to discover products and not buy them.
How ready are fashion shoppers for autonomous AI purchase?
Just 7 per cent of respondents would allow an AI assistant to purchase with some degree of autonomy, either within a predefined budget (5 per cent) or for certain retailers only (2 per cent). ACI Worldwide's research found that 53 per cent of fashion shoppers are unwilling to let AI purchase on their behalf under any circumstances, and only 7 per cent would allow some degree of autonomous purchasing, usually within a set budget or for specific retailers. The New Rules of Fashion and Sportswear Payments draws on a YouGov survey of 3,000 US and UK shoppers conducted in June 2026.
Retail was among the more difficult sectors to crack when it comes to AI agents, the report found. Seventy-eight per cent of retail respondents were found to be at least somewhat open to an AI agent being involved in their shopping, compared with 91 per cent for travel.
For now, fashion and home furnishings appear less suited to fully autonomous agentic commerce. In these segments, AI is therefore more likely to serve as a decision-support tool than as a fully autonomous purchasing agent. Kearney estimates that 73 per cent of consumers are familiar with AI tools and 60 per cent expect to use AI agents for shopping within the next 12 months, signalling a near-term inflection in intent.
Frequently asked questions
What is the difference between AI shopping assistants and agentic commerce?
AI shopping assistants typically provide recommendations and answer questions but require humans to complete checkout manually. Agentic commerce describes systems where AI agents can execute part or all of the purchase process autonomously, from product discovery through payment, with varying levels of human oversight depending on the use case and user preferences.
Which fashion retailers have deployed agentic shopping features?
Amazon's Rufus assistant reached 300 million customers in 2025. Daydream launched an autonomous fashion shopping agent in November 2026 that can style outfits and complete purchases with merchant partners. Lowe's and The Home Depot introduced agentic apps in March 2025. Fashion brands can also enable discovery through platforms like ChatGPT, Google AI Mode and Perplexity, though most shoppers currently use these for research rather than autonomous purchase.
Are fashion consumers ready to let AI agents buy clothes autonomously?
Only 7 per cent of fashion shoppers would allow an AI assistant to purchase with some degree of autonomy, according to a June 2026 survey of 3,000 US and UK consumers. Fifty-three per cent are unwilling to let AI purchase on their behalf under any circumstances. However, 85 per cent of luxury consumers use AI assistants to support shopping decisions, and 60 per cent of consumers expect to use AI agents for shopping within the next 12 months, suggesting interest in assisted rather than fully autonomous purchase.
Sources
- McKinsey: Who wins when AI does the shopping? New evidence on agentic commerce
- McKinsey: The State of Fashion 2026
- Business of Fashion: AI's Transformation of Online Shopping Is Just Getting Started
- ACI Worldwide: Agentic commerce in fashion: Shoppers want AI to save them money, not spend it
- Shopify: Form 10-K FY2025
- PRNewswire: Daydream Launches First-Ever Autonomous Agent for Fashion Shopping, Styling and Checkout
